Discuss features as they are added to the new version. Give us your feedback. Don't post bug reports, feature requests, support questions or suggestions here.
Forum rules
Discuss features as they are added to the new version. Give us your feedback. Don't post bug reports, feature requests, support questions or suggestions here. Feature requests are closed.
Schumi wrote:
Lastof meant http://www.bokt.nl . Although I don't know if they are realy on the same server - at least area51 and bokt.nl don't have the same IP but this could still mean that they run virtual hosts on the same machine.
8O
Total posts: 21004670, total threads: 681012 and 62595 registered users!
Schumi wrote:
Lastof meant http://www.bokt.nl . Although I don't know if they are realy on the same server - at least area51 and bokt.nl don't have the same IP but this could still mean that they run virtual hosts on the same machine.
8O
Total posts: 21004670, total threads: 681012 and 62595 registered users!
I would probably run that large of a forum in debug mode as well, so you can catch any errors before they cause much havoc.
It would be a real pain if something went wrong with that large of a database.
My phpBB3 Mods || My Mod Queue Search Engine Friendly (SEO) URLs || Profile link on Avatar and/or Username || AJAX Chat
Display Posts Anywhere || CashMod || AJAX Quick Edit || AJAX Quick Reply
So far it seems to be... but I don't have a 1 vs 2 test to verify any stats or presumptions.
It is supposed to handle caching more efficiently, plus it's coded more efficiently... all of that ads up to less load on the server.
So, I am unable to offer you cement proof, I don't have the time to go perform the test to give you the statistics, but it is supposed to be lighter on the server.
if you have a half-decent template (and subsilver is way better than half-decent), the main factor of the bbs performance is the db back end.
as a result, the main factor of the bb performance is the number and complexity of the queries.
i didn't actually measure it, but it is clear that olympus uses way less queries than phpbb 2.0 for all the main tasks. in some cases it seems that it uses less than half the queries.
as a result, it runs faster while using less cpu, disk and memory resources.
in addition, the template system, which only compiles the new template file when the source file (ie., the .html) is changed, requires much less cpu and memory resources (for a negligible price in disk space).
even better, by getting rid of all the eval() calls that the old (2.0.x) templating system used, the code is much more compatible with many external php optimizers (which are very widely used). it is very difficult to optimize eval() calls.
so, all in all, without any measurements, olympus should be roughly twice as efficient as phpbb 2.0.x (or slightly better) where no php optimizer is used, and the gap in performance widens with such an optimizer.
as to the sql-caching scheme: i don't know of any measurements anyone have done to test it. personally, i wish they would have built a way to disable the sql-caching, so we could experiment with it (i have some suspicion that the sql-caching might be, in some circumstances, counterproductive, but unfortunately there is no simple way to test it, afaik).